Hi,
I'm trying to get confidence score for each word in the extracted content. I noticed character-variants score on using "xml:writeRecognitionVariants" parameter. But it would be better if i could able to get scores on word basis. Any possible ways?
Thanks,
vishnu
コメント
1件のコメント
Hi Vishnu,
It is possible to get the confidence level for the character only. Despite the fact that there is no option to get confidence level for a word using Cloud OCR SDK, you can estimate it on your side using the output XML document.
Hope this information will be helpful!
サインインしてコメントを残してください。